Search

Age-associated B cells (ABCs) represent a distinct cell population expressing low levels of CD21 (CD21 −/low ). The Ig repertoire expressed by ABCs in aged mice is diverse and exhib its signs of somatic hypermutation (SHM). A CD21 −/low B-cell population is expanded in autoimmune diseases, e.g. systemic lupus erythematosus, as well as in lupus-prone NZB/W mice and in mice lacking a pre-B cell receptor (SLC −/− ). However, the nature of the CD21 −/low B cells (hereafter ABCs) in autoimmunity is not well understood. Here we show that in young SLC −/− mice, the vast majority of the ABCs express memory B-cell (MBC) markers in contrast to wild-type controls. A similar population is present in lupus-prone MRL mice before and at disease onset. In SLC −/− mice, a majority of the ABCs are IgM + , their V H genes have undergone SHM, show clonal diversification and clonal restriction at the H-CDR3 level. ABC hybridomas, established from SLC −/− mice, secrete typical lupus autoantibodies, e.g. anti-Smith antigen, and some of those that bind to DNA comprise a H-CDR3 that is identical to previously described IgM anti-DNA antibodies from lupus-prone mice. Together, these results reveal that ABCs in autoimmune mice are comprised of autoreactive MBCs expressing highly restricted H-CDR3 repertoires.

Mice lacking C3, C4 or complement receptor 1/2 (Cr) have defective germinal centers (GC). The requirement for C4 implicates complement fixation by immune complexes (IC) via the classical pathway. Yet, transgenic (Tg) mice that lack circulating antibody but still express membrane IgM (mIgM) have normal GC responses. We showed previously that cross-linking mIgM leads to the deposition of C3 on the B cell surface and that disruption of this pathway diminishes GC responses. Here, we investigate the role of Cr in this process by generating mIgM-Tg mice that lack Cr and serum Ig. These mIgM/Cr-/- mice have smaller, transient GC, with incomplete B cell receptor down-regulation and peanut agglutinin up-regulation, compared to mIgM/Crwt counterparts. BM chimera experiments showed that Cr on B cells is required for normal GC responses. These results establish that Cr ligands generated at the B cell surface are sufficient for normal GC responses and function by signaling Cr on B cells. Unexpectedly, chimera experiments also showed a critical role for Cr on follicular dendritic cells (FDC), even in the absence of IC, indicating novel functions for FDC-expressed Cr beyond the capture of C3-coated IC.

Microparticles are small vesicles released from the plasma membrane of various cell types independently of apoptosis or cell death, are transferred between cells, and carry membrane proteins from one cell to another. We have studied the mechanism of uptake of microparticles by monocytes and B cells. The transfer of microparticles to B cells was almost completely dependent on complement. Incubation of microparticles with serum resulted in opsonization of microparticles with the complement cleavage product iC3b. The subsequent transfer to B cells was mediated by the complement receptor CR2. The interaction between iC3b-opsonized microparticles and B cells reduced the activation of B cells as measured by expression of MHC class II, CD86 and CD25. In contrast, transfer of microparticles to monocytes was only partially complement dependent, but involved calcium and annexin V, and was found to change the cytokine profile of monocytes towards a reduced release of the pro-inflammatory cytokines GM-CSF and TNF-alpha and an increased release of the anti-inflammatory cytokine IL-10. These data show that microparticles are taken up by B cells and monocytes by different mechanisms and modulate the activation of monocytes and B cells towards an anti-inflammatory phenotype. Microparticles might be involved in counterbalancing pro-inflammatory signals arising from tissue injury or inflammation.

Allergic diseases are characterized by the presence of eosinophils, which are recruited to the affected tissues by chemoattractants produced by T cells, mast cells and epithelium. Our objective was to evaluate if allergens can directly activate human eosinophils. The capacity of purified allergen extracts to elicit eosinophil chemotaxis, respiratory burst, degranulation and up-regulation of the adhesion molecule complement receptor 3 (CR3) was determined in eosinophils isolated from healthy blood donors. Eosinophils stimulated with an extract from house dust mite (HDM) released the granule protein major basic protein (MBP) and up-regulated the surface expression of CR3. Cat allergen extracts also induced the up-regulation of CR3, but not the release of MBP; instead cat, as well as birch and grass allergens, elicited the release of eosinophil peroxidase (EPO). In addition, grass pollen extract caused the secretion of MBP. None of the allergens stimulated eosinophilic cationic protein release, nor production of free oxygen radicals. Both HDM and birch extracts were chemotactic for eosinophils. These findings establish that common aeroallergens can directly activate eosinophils in vitro. We propose that eosinophil activation in vivo is not exclusively mediated by cytokines and chemokines of the allergic inflammatory reaction, but could partly be the result of direct interaction between allergens and eosinophils.

The C3b-binding receptor, CR1/CD35, supports CR2/CD21-mediated activation of complement by human B lymphocytes, possibly by associating with CR2 to promote or stabilize the binding of hydrolyzed C3 (C3i), the primary component of the AP convertase, C3i-Bb. To evaluate this hypothesis, we examined the uptake kinetics and binding equilibria for C3i dimer interaction with human blood cells in the absence and presence of CR1- and CR2-blocking mAb. C3i displayed dual uptake kinetics to B lymphocytes, comprising of rapid binding to CR1 and slower binding to CR2. The forward rate constants (k(1)) for CR1 and CR2, operating independently, differed ca. 9-fold (k(1)=193+/-9.4 and 22.2+/-6.0 x 10(3) M(-1)s(-1), respectively). Equilibrium binding of C3i to B lymphocytes was also complex, varying in strength by ca. 13-fold over the C3i concentration range examined. The maximum association constant (K(a, max)=109+/-27.2 x 10(7) l/mole) was ca. 9- and 6-fold greater, respectively, than those for CR1 or CR2 acting alone (K(a)=13.2+/-5.3 and 18.5+/-3.5 x 10(7) l/mole). The high avidity of the CR1-CR2 complex for C3i is consistent with its rates of C3i uptake and release being determined by CR1 and CR2, respectively.

We previously demonstrated that CR2 activation on human B lymphocyte surface specifically triggered tyrosine phosphorylation of the 95-kDa nucleolin, this leading to its binding on SH2 domains of p85 sub-unit of PI 3-kinase and to activation of this enzyme. The specificity of CR2 pathway was clearly demonstrated as neither CD19 nor BCR could induce tyrosine phosphorylation of nucleolin in normal B lymphocytes. These data led us to investigate herein additional molecular events, which were triggered by CR2 activation, upstream and downstream to PI 3-kinase activation. Upstream, we demonstrated that pp60src, a tyrosine kinase of the src family, was involved in tyrosine phosphorylation of nucleolin, while syk tyrosine kinase was not. We also demonstrated a direct protein-protein interaction of pp60src with nucleolin in a CR2-dependent and CD19-independent pathway. Downstream, we demonstrated that CR2 activation also triggered Akt and GSK3 enzyme activation, this pathway being under the control of pp60src tyrosine kinase activation. These regulatory functions of activated CR2 were specific as independent of syk tyrosine kinase and of CD19 and BCR activation. Thus, CR2 activation recruits a specific mechanism to activate PI 3-kinase and its subsequent pathways, this mechanism being different to those recruited by CD19 and BCR.

Epitope density and organization have been shown to be important factors for B cell activation in many animal model systems. However, it has been difficult to separate the role of antigen organization from the role of local antigen concentrations because highly organized antigens are usually particulate whereas non-organized antigens are more soluble. Hence, highly organized and non-organized antigens may interact with different cell types and in different locations within lymphoid organs. In order to assess the role of antigen organization in regulating B cell responses, we immunized mice with highly repetitive virus-like particles, which exhibit different epitope densities covalently attached to them. Therefore, the same particulate structure was used to present identical epitopes that differed in their degree of organization. Induction of epitope-specific IgM titers, reflecting early B cell activation, were unaffected by the degree of epitope density. Furthermore, the absence of Th cells or CD21/CD35 did not reduce the IgM response. In contrast, the degree of organization was a critical factor influencing the magnitude of the epitope-specific IgG response. Moreover, the threshold for IgG responses was shifted in the absence of CD21/CD35, resulting in the requirement for higher epitope densities to allow efficient IgG responses. Thus, IgG but not IgM responses are regulated by epitope density and B cell costimulatory thresholds.

A specialized subpopulation of T lymphocytes is reactive to the MHC class I-like molecule CD1d. It is not clear which cells are the major antigen-presenting cells in vivo in the activation of CD1-restricted immune responses. We have characterized a subset of B lymphocytes expressing six- to eightfold higher levels of CD1 than the bulk of B cells. The cells have a surface phenotype (CD21(high), CD23(low), IgM(high), IgD(low)) found previously to characterize B cells residing in the splenic marginal zones. CD1(high) B cells localize preferentially to the spleen, and appear late in ontogeny, at 3 - 4 weeks of age. The CD1(high) B cells were present in mice lacking conventional helper T cells, ruling out an exclusive origin from T cell-dependent immune responses. Still, some CD1(high) B cells had been involved in T cell-dependent immune responses as suggested by mutations in their rearranged immunoglobulin gene regions. The population could still be found in mice with severely reduced B cell reactivity to bacterial lipopplysaccharides (C3H / HeJ mice) and in mice unable to respond to thymus-independent type 2 antigens (NFR.Xid mice), as well as in germ-free mice, indicating that bacterial antigens are not major stimuli for the induction of CD1(high) B cells. In contrast, the CD1(high) B cell population was severely reduced in CD19-deficient mice. Taken together, the results imply that the CD1(high) population is heterogenous and of mixed origin, dependent for its development or maintenance on signaling through the CD19 molecule.

Der p I, a cysteine protease representing a major allergen of the house dust mite Dermatophagoides pteronyssinus, has recently been shown to cleave CD23 from the surface of cultured human B cells (RPMI 8866 B cell line). We have now undertaken a detailed investigation of CD23 cleavage by Der p I. We demonstrate that Der p I cleaves CD23 at two sites (Ser155-Ser156 and Glu298-Ser299) to produce a 17-kDa fragment containing the lectin domain and only part of the C-terminal tail. No such effect was demonstrable with mouse CD23, a finding which was anticipated based on its lack of the cleavage sites identified on human CD23. Based on the cleavage pattern and the model of CD23, we propose a sequence of events leading to the liberation of the 17-kDa soluble CD23 fragment. The biological significance of such cleavage is underlined by the demonstration that Der p I-treated B lymphocytes lose their ability to bind IgE, and that the 17-kDa fragment (amino acids 156-298) contains the minimum structural requirement (amino acids 156-288) for binding to both IgE and CD21.

CD21 (complement receptor 2, CR2) binds the C3 degradation products, iC3b and C3d, which are covalently linked to antigen or immune complexes in the process of complement activation. The ability of antigen-nonspecific B cells to present immune complexes containing high titers of acquired antibodies was tested. Influenza virus was incubated with serum from immune donors to create complement-containing complexes. These bound specifically to CD21 on transfected fibroblasts and B cell lines, as measured by microcytofluorimetry. Binding of immune complexes was ablated by inactivation of serum complement. In addition, the immunoglobulin in immune human serum blocked influenza binding to cells in the absence of complement, implying a minimal role for immunoglobulin-Fc receptor interactions in this system. Significant immune complex binding required a threshold level of CD21 expression, suggesting that only those cells with the highest levels of CD21 are likely to participate in the processing of macromolecular antigens. B cells pulsed with complement-influenza complexes elicited an augmented response from a panel of influenza-specific, class II-restricted T cell clones, as compared with those which had bound immunoglobulin-influenza complexes lacking complement. This enhanced response did not require CD35. In addition, B cell lines expressing higher levels of CD21 were more efficient in processing antigen than those with lower levels. These findings suggest that presentation of antigen by B cells in immune individuals is dependent on the binding of complement-antigen immune complexes to CD21.

Stimulation in vitro of murine splenic B cells by lipopolysaccharide, anti-kappa Sepharose, anti-CD40 or allo-reactive T helper cells all up-regulated CD21 and CD23 surface expression. Neither anti-CD21 nor anti-CD23 antibodies induced B cell growth or differentiation when added in soluble form or coupled to Sepharose. However, anti-CD40-stimulated B cells showed increased proliferation in the presence of anti-CD21 antibodies coupled to Sepharose; co-stimulation via CD21 also induced differentiation to immunoglobulin secretion in a fraction of anti-CD40-stimulated B cells. Furthermore, anti-CD40 antibodies inhibited differentiation to immunoglobulin secretion induced by lipopolysaccharide and, hence, appears to be a dominant negative signal for B cell differentiation.

We report on a soluble (s) form of CD21 (the C3dg/Epstein-Barr virus receptor, CR2) that is spontaneously released by B and T lymphocytes. Immunoprecipitation with anti-CD21 mAb of culture supernatants of surface and biosynthetically labeled B and T cell lines revealed a single band with an apparent molecular mass of 135 kDa. The molecule exhibited a molecular mass 10 kDa lower than that of membrane CD21. The release of soluble CD21 (sCD21) was time dependent and correlated with a parallel decrease in the expression of the membrane-associated molecule. The protein was also found in culture supernatants of tonsillar B cells and normal human thymocytes. Epitopic analysis using combinations of anti-CD21 monoclonal antibodies (mAb) indicated that sCD21 and membrane CD21 were similarly recognized by mAb directed against short consensus repeats (SCR) 1-2, SCR 4-5 and SCR 9-11. Affinity-purified sCD21 was capable of binding to purified human iC3b and to human recombinant CD23, as assessed by enzyme-linked immunosorbent assay and by using the BIAcore technology. In addition, normal human serum was found to contain a soluble form of CD21 that exhibited a similar molecular mass to that of the molecule shed by B and T cells in culture. The serum form of CD21 was recognized by all anti-CD21 mAb that we tested and showed a high reactivity with mAb directed against SCR 1-2. Our observations suggest that B and T cells shed the extracellular portion of CD21 and release a soluble molecule that retains the ligand-binding properties of CD21, thus having a potential role in immunoregulation.

Previous studies have documented a role for membrane-bound CD23 (the low affinity Fc epsilon RII) in presentation of alloantigens by B cells. The aim of the present study was to examine the involvement of cell surface CD23 in presentation of more conventional soluble protein antigens to T cells. We show that antibodies to CD23 and to its lymphocyte-associated second ligand, CD21, inhibit presentation of the cow's milk allergen casein, by autologous CD23+CD21+ B-EBV cell lines to casein-specific HLA-DP-restricted CD4+ T cell clones obtained from patients with either reaginic or enterophatic forms of cow's milk protein intolerance. Maximal inhibition was achieved when the antibodies were added at the initiation of the culture. The absence of specific inhibition by an anti-DR alpha monoclonal antibody (mAb) argues against a steric hindrance phenomenon impeding access of the T cell receptor to major histocompatibility complex class II molecules. Rather, anti-CD23 and anti-CD21 mAb-induced inhibition of antigen presentation seems to affect at least partly, heterotypic conjugate formation through CD23/CD21 interaction. Double immunofluorescence labeling of the T cell clones and antibody inhibition of T/B conjugate formation shows that functional CD23 and CD21 molecules are induced on T cells following contact with B-EBV cell lines. Taken together, these data indicate that CD23/CD21 interactions between T and B cells are required for presentation of soluble protein antigens by B-EBV cell lines to specific CD4+ T cells. The potential implications of these findings for allergen-specific T cell activation are discussed.Previous studies have documented a role for membrane-bound CD23 (the low affinity Fc epsilon RII) in presentation of alloantigens by B cells. The aim of the present study was to examine the involvement of cell surface CD23 in presentation of more conventional soluble protein antigens to T cells. We show that antibodies to CD23 and to its lymphocyte-associated second ligand, CD21, inhibit presentation of the cow's milk allergen casein, by autologous CD23+CD21+ B-EBV cell lines to casein-specific HLA-DP-restricted CD4+ T cell clones obtained from patients with either reaginic or enterophatic forms of cow's milk protein intolerance. Maximal inhibition was achieved when the antibodies were added at the initiation of the culture. The absence of specific inhibition by an anti-DR alpha monoclonal antibody (mAb) argues against a steric hindrance phenomenon impeding access of the T cell receptor to major histocompatibility complex class II molecules. Rather, anti-CD23 and anti-CD21 mAb-induced inhibition of antigen presentation seems to affect at least partly, heterotypic conjugate formation through CD23/CD21 interaction. Double immunofluorescence labeling of the T cell clones and antibody inhibition of T/B conjugate formation shows that functional CD23 and CD21 molecules are induced on T cells following contact with B-EBV cell lines. Taken together, these data indicate that CD23/CD21 interactions between T and B cells are required for presentation of soluble protein antigens by B-EBV cell lines to specific CD4+ T cells. The potential implications of these findings for allergen-specific T cell activation are discussed.

The present study demonstrates that the C3b receptor CR1 (CD35) and the C3dg/Epstein-Barr virus receptor CR2 (CD21) are expressed by 25% and 70% of normal human thymocytes, respectively. The expression of CR2 extends to both CD1+ and CD1- cells in the thymus. Two subsets of CR2+ thymocytes were defined expressing low and high density of the receptor. The CR2++ subset represented 20% of CR2+ thymocytes and co-expressed the CR1 receptor. CR2++ thymocytes expressed an immature CD1dull, CD3-, CD4dull, CD8-, CD7++ phenotype and included a subpopulation of large cells expressing CD34. Twenty percent of thymocytes expressed the CD21 epitope defined by monoclonal antibody BU32, which is involved in the binding of CD23 to CD21. These observations provide a basis for a role for CD21 in the proliferation and differentiation of thymocytes at early stages of maturation. The functionality of CR1 and CR2 on thymocytes was evidenced by the ability of the receptors to mediate infection of cells with complement-opsonized human immunodeficiency virus (HIV). The results may be relevant to the immunopathogenesis of HIV infection.

Murine macrophages have been described as possessing complement receptors for C3b and iC3b. These binding activities have been assumed to be due to the presence of the CR1 and CR3 proteins, respectively. The mouse Cr2 gene produces two distinct gene products of approximately 190,000 relative molecular mass (M(r)) (Cr2-190) and 145,000 M(r) (Cr2-145). Because of the similarity in size to human complement receptors, the Cr2-190 protein has been dubbed murine CR1 while the murine Cr2-145 product has been termed murine CR2. In order to define the complement receptor genes expressed by murine macrophages, we investigated the expression patterns of Cr2-190, Cr2-145 and another mouse complement receptor, Crry, in three different mouse macrophage populations: bone marrow-derived macrophages, thioglycollate-elicited peritoneal macrophages and the macrophage cell line, J774. Neither of the Cr2 gene transcripts encoding the Cr2-145 and Cr2-190 proteins could be detected in these populations by RT-RPCR analysis although Crry transcripts were evident. Cr2-145 and Cr2s-190 proteins could not be detected on the surface of thioglycollate-elicited macrophages using a monoclonal antibody that recognizes both proteins. Thus, contrary to previously published data, murine macrophages do not possess the Cr2 gene products.

The expression of complement receptor 2 (CR2) has been demonstrated in established HTLV-1-transformed cell lines and in 12 studied de novo infected peripheral blood lymphocytes cultures, using 2 HTLV-1 sources. The simultaneous detection of CR2 and HTLV-1 antigens in both co-cultivated and supernatant-infected peripheral blood lymphocytes suggest that the increased CR2 expression is in tandem with the increasing HTLV-1 antigen expression. CR2 up-regulation seen during polyclonal activation is presumably in response to a viral protein, although a cellular factor has not been ruled out. Increasing CR2 expression during early infection suggests its possible involvement in selection or development of subsequent transformation events. Variable levels of CR2 in immortalized cell lines argue against its obligate expression of function in the maintenance of the transformed state. The expression of CR2 in cellular activation of T cells may be stage restricted. This study also expands the cellular distribution for CR2.

Germinal center cells (GCC) are programmed to die by apoptosis unless they receive a positive signal for rescue. The primary signal in vivo is believed to be dependent on interaction with antigen held as immune complexes on follicular dendritic cells (FDC), a subset of which express large amounts of CD23, a low-affinity receptor for IgE. Recombinant soluble CD23 (sCD23) and interleukin-1 have been found to potentiate the survival of GCC in vitro. Recently, CD23 was shown to interact specifically with a ligand other than IgE, namely CD21 (CR2/Epstein-Barr virus receptor). In the present study, we show that a subset of anti-CD21 monoclonal antibodies behave similarly to soluble CD23 in their effect on GCC inasmuch as they: (i) diminish the occurrence of apoptosis; (ii) promote a plasmacytoid appearance in rescued cells; (iii) up-regulate expression of the Bcl-2 proto-oncogene. These findings indicate that FDC-derived CD23 exerts its effects on GCC via CD21.

Epstein-Barr virus (EBV), the causative agent of mononucleosis and several human cancers, infects cells via complement receptor type 2 (CR2, CD21) which also serves as the receptor for the third complement component, C3. Expression of this receptor is restricted to B lymphocytes, immature thymocytes, and certain epithelial cells. In the present investigation; we describe the presence of a seemingly novel EBV receptor which is phenotypically distinct from CR2. Among various leukemic T cells studied, one, HSB-2, demonstrates no reactivity to several anti-CR2 antibodies, yet it reacts strongly with EBV as detected by incubation with biotin-conjugated virus and streptavidin-phycoerythrin. The virus binding is specific as demonstrated by blocking with anti-EBV antibodies and with non-conjugated virus. Aggregated C3 also binds HSB-2 and is capable of partially inhibiting EBV binding. The absence of CR2 on HSB-2 is further supported by the lack of expression of specific mRNA, assessed by Northern blotting analysis and polymerase chain reaction. Viral internalization and infection is demonstrated with electron microscopy, with detection of EBV-DNA by Southern blotting, and with detection of EBNA-1 transcripts by the polymerase chain reaction. Even though HSB-2 does not express CR2, it nevertheless displays transcripts which have some homology to a CR2 cDNA probe under low stringency hybridization conditions. This probe encompasses approximately the N-terminal half of CR2 which includes the EBV-binding epitope(s). The HSB-2 message is 5.2 kb, a size distinct from the 4.7-kb message of B cell CR2s. In contrast, the 5.2-kb message in not seen, under similar hybridization conditions, with a probe comprising the C-terminal half of CR2. Collectively, the data indicate that a receptor molecule having distinct phenotypic characteristics from the known CR2 protein on B cells is utilized by EBV to target human T lymphocytes.

Tumor necrosis factor-α (TNF-α), which is produced mainly by monocyte/macrophage cells, has diverse physiological functions on lymphoid cells. Moreover, it has been shown that TNF-α exhibits antiviral activities. Here we report that Epstein-Barr virus (EBV), a B lymphotropic human herpes virus that interacts intimately with the immune system, exerts a strong inhibitory effect on TNF-α production by lipopolysaccharide-treated peripheral blood leukocytes as well as by monocytic cell lines, HL-60 and U-937. Flow cytometric analysis following staining with OKB7 monoclonal antibody showed that about 20% of cells from these monocytic lines express the CR2 antigen. Direct binding of fluorescein isothiocyanate-labeled EBV indicated that the virus binds to approximately 22% of cells of both monocytic lines. However, no virus-specific antigens were detected in the infected cells by immunofluorescence, suggesting that the infection was of the abortive type. The use of UV- or heat-inactivated EBV and EBV treated with neutralizing antibodies resulted in the abrogation of the EBV inhibitory effect on TNF-α synthesis. These results suggest that infectious virus is necessary to obtain such an inhibitory effect. Analysis of TNF-α mRNA by polymerase chain reaction amplification indicated that the EBV suppressive effect is manifested at the transcriptional level. In contrast, EBV did not inhibit interleukin 1 mRNA production by these cells. These results indicate that EBV interacts directly with monocytes/macrophages to exert its immunomodulatory effect.

Complement receptor type II (CR2/CD21) is the major receptor for C3d fragments on immune complexes. CD21 also serves as the receptor for Epstein-Barr virus in humans. On mature B cells, CD21 reduces the threshold of BCR signaling together with CD81, Leu13 and CD19, but it also occurs on other cells of the immune system where it performs unknown functions. A soluble form of CD21 (sCD21) is shed from the cell surface and is found in human blood plasma. An as-yet-unknown protease is thought to be responsible for this shedding. Altered levels of sCD21 occur in plasma in certain clinical conditions. We show here by mass spectrometry that sCD21 in human plasma of healthy donors is predominantly a short form of CD21 without the exon-11-encoded sequences. Whereas the N terminus of sCD21 was found unmodified, the C terminus is truncated, implying that only the extracellular portion of CD21 is shed. Peripheral blood B cells, but not T cells, contribute to the plasma CD21-pool. CD21 shedding is induced by stimulation with PMA plus Ca(2+) ionophore, or by stimulation of the BCR with anti-IgM+anti-CD40.

Resting CD4(+) T cells in the lymphoid tissue (LT) are essential producers of virions at the beginning of HIV infection in vivo. We previously developed a model that allowed in vitro infection of non-prestimulated T lymphocytes in the presence of autologous B lymphocytes and complement. In this study, we try to clarify the mechanism(s) responsible for virus transmission in unstimulated autologous B cell/T cell co-cultures. Ex vivo analyses of patient plasma samples revealed that HIV was opsonized. Flow cytometry showed that opsonized virus preferentially bound to complement receptor (CR)-2 on B lymphocytes in primary B cell/T cell co-cultures. As indicated by cytokine measurements and transwell experiments, soluble factors seemed to play a minor role in enabling infection. Rather, direct interaction between B and T lymphocytes and direct binding of opsonized virus to CR2 on B cells turned out to be essential for productive infection. Antibodies blocking cell-cell adhesion inhibited p24 antigen production. An anti-CR2 antibody blocking C3d-CR2 binding also significantly reduced viral replication. Since the infection of unstimulated T cells by opsonized primary HIV isolates in the presence of B cells was highly efficient independent of the tropism of the virus, this mechanism may be critical in the pathogenesis of HIV.

Morphological and kinetic studies of immune complex (IC) trapping by follicular dendritic cells (FDC) show marked age-related deficits. We postulated that a reduction in trapped IC, which generate CD21 ligands (L) on FDC, would lead to inadequate FDC-Ag-B cell interactions resulting in depressed Ab responses. To determine whether the age-related defect was the result of the aging of FDC or changes in the in vivo microenvironment of FDC (i.e. aging B and T cells), FDC-B cell-T cell-Ag interactions were studied in in vitro germinal centers where various combinations of old and young cells could be compared. Since we reasoned that reduced IC on FDC would generate less CD21L needed to stimulate the B cell co-receptor via CD21, we also examined the role of complement (C'). The hypothesis that aging reduces the accessory activity of FDC was tested with increasing numbers of FDC from young (12 weeks) or old (20 months) mice in the presence of young (12 weeks) B and T lymphocytes. The Ag-specific stimulatory activity of FDC was studied using the OVA-specific Ab response which was reduced by 40-50% in the presence of old FDC. Antigen-independent FDC-mediated co-stimulation was studied by using LPS to stimulate B-lymphocytes to produce immunoglobulin (Ig). In the presence of old FDC, co-stimulation was decreased by 70-80% in the LPS system. Incubation of aged FDC with IC and C' to provide FDC with CD21L restored co-stimulatory activity to near normal levels. In marked contrast, no defects in old B and T cells were apparent. The data suggest that the Ag handling capacity and co-stimulatory activity of old FDC become defective with aging and this appears to be a consequence of reduced trapping and presentation FDC-Ag and CD21L to B cells.

The K/BxN murine model of rheumatoid arthritis (RA) is dependent on the specificity of the KRN alpha beta-TCR, to recognize glucose-6-phosphate-isomerase (GPI) on the NOD MHC class II A(g7) allele and production of GPI-specific autoantibodies. Transfer of K/BxN serum into MHC-unrelated and lymphocyte-deficient mice induces RA. To investigate whether K/BxN serum-induced RA involves complement activation and/or the complement receptors (CR) 1 and 2, we analyzed the role of complement C4 and of CR1 and CR2. For this purpose we used C4(-/-) mice impaired in the classical and the lectin complement pathways; Cr2(-/-) mice lacking CR1 and CR2 and, as control strains, BALB/c, C57BL/6, KRN and NOD. RA was assessed by caliper measurement of ankle thickness, clinical index and joint histology. We found that all mouse strains except NOD developed RA. The lack of protection in C4(-/-) mice suggests that antibody-mediated RA is independent of the classical as well as the lectin complement pathways and the split complement product C4b. The lack of protection in Cr2(-/-) mice suggests that absence of CR1 had no significant affect, considering its role in immune complex clearance, inhibition of C3 and C5 convertase and as receptor for C3b/C4b. Also, CR2 lacks a role in disease as analyzed here, in its possible functions as receptor for C3dg, germinal center reaction and activation of alternative pathway on binding iC3. Hence we conclude that the transmission of K/BxN serum-induced RA is independent of the classical and the lectin complement pathways and CR1 and CR2. The crucial role of complement C5, while neither classical nor lectin pathway is necessary, indicates that the alternative complement pathway may have a role in the K/BxN serum-induced RA model.

Productive follicular dendritic cell (FDC)-B cell interactions appear to involve critical ligand-receptor interactions. Immune complexes (IC) on FDC activate complement and provide FDC with a complement-derived CD21 ligand (CD21L), which bind CD21, while antigen in the IC binds on the B cell-BCR. Further, FDC-FcgammaRIIB binds Fc regions of antibodies in IC and reduces coligation of BCR and FcgammaRIIB minimizing an inhibitor of B cell activation. Given that Fc receptors and complement receptors bind immunoglobulins and complement fragments of other species, we reasoned that FDC accessory activity should cross MHC and species barriers. This prediction was tested using memory lymphocytes from OVA-immune mice and TT-immune humans in combination with FDC from murine lymph nodes and human tonsils. Human and murine FDC converted IC into potent immunogens (specific antibody increased from background to thousands of ng / ml). MHC barriers did not restrict this activity and human FDC worked with murine lymphocytes and murine FDC worked with human lymphocytes. Furthermore, stimulation via MHC-dependent allogeneic or zenogeneic mechanisms did not promote antibody production by FDC. Polyclonal responses stimulated by lipopolysaccharide and pokeweed mitogen were also promoted (10 - 100-fold) and anti-CD21 blocked FDC activity. These results substantiate the hypothesis that FDC are necessary for strong recall responses and that FDC-CD21L is critical.

We coupled a monoclonal anti-human IgD to the gp350 gylcoprotein of Epstein-Barr virus, which has been shown to bind to the complement receptor 2 (CR2), and compared its B cell stimulatory ability to that of anti-Ig and to a multivalent anti-Ig-dextran conjugate. The anti-Ig-gp350 conjugate stimulated higher levels of human B cell proliferation in vitro than did anti-Ig or anti-Ig conjugated to control viral protein, comparable to the proliferation stimulated by the multivalent anti-Ig-dextran. This enhanced proliferation was dependent on binding of the conjugate to CR2, inasmuch as an anti-CD2 antibody blocked the enhanced proliferative response. This enhanced proliferative response was associated with prolonged elevations of intracellular ionized calcium, which was comparable to the response stimulated by anti-Ig-dextran. These findings suggest the use of gp350 as a carrier molecule for weakly immunogenic peptides or antigens which, when bound to gp350, would enhance B cell clonal expansion and activation of antigen-specific B cells.

The role of complement receptor 2 (CR2) short consensus repeats (SCR) in binding of hydrolyzed C3 (iC3) to form an alternative pathway (AP) convertase, and promoting C3 fragment deposition following AP activation, was examined. We used (1) K562 cells transfected with CR2 constructs, where the C3d-binding site of CR2 (SCR1+2) was replaced with the four-SCR vaccinia virus complement control protein (VCP), or truncation mutants thereof, and (2) COS cells transfected with wild-type (wt) CR2, or deletion mutants thereof. AP activation required iC3 binding in both systems. Thus, the VCP-CR2 chimera had an iC3 binding efficiency of 11.4 %, compared to wtCR2, and a relative AP activity of 5.5 %, the truncation mutants being inactive. Of the CR2 mutants, only EK (DeltaSCR10 - 11) had AP activity similar to wtCR2. NN (DeltaSCR6 - 8) and NOP (DeltaSCR6-mid14) had reduced AP activity, but near normal iC3 binding. XB (DeltaSCR3 - 6) and PP (DeltaSCR3-mid14) were inactive in both assays. We conclude that, whilst iC3 binding to CR2 via SCR1 - 4 is essential for AP activation, the efficiency of C3 deposition also depends on the midportion of CR2.

Interactions between CD23, the low-affinity receptor for IgE, and CD21, the C3d/EBV receptor, modulate several intracellular events in lymphocytes. A soluble form of CD21 (sCD21) corresponding to the extracellular domain of the receptor circulates in normal plasma. We now demonstrate that purified sCD21 acts as a functional ligand for CD23-expressing monocytes. Soluble CD21 induced an increase in intracellular cGMP levels and the production of IL-6 and TNF-alpha in IL-4-pretreated monocytes induced to express CD23 but not in unstimulated CD23- monocytes. The accumulation of cGMP and the production of TNF-alpha were inhibited by NG-monomethyl-L-arginine (L-NMMA), indicating that sCD21 activates the L-arginine pathway of NO production. We demonstrated that sCD21 activates NO synthase (NOS) since it was found to enhance the conversion of L-arginine into L-citrulline and induce the intracellular expression of inducible NOS in CD23+ monocytes. In addition, sCD21 was shown to up-regulate the expression of HLA-DR and CD40 and decrease that of CD14 on cultured CD23+ monocytes. Thus, in a fashion similar to IgE complexes, sCD21 is able to efficiently trigger CD23 signaling pathways, inducing the release of pro-inflammatory mediators by human monocytes. Soluble CD21 up-regulates the expression of molecules involved in antigen presentation, further suggesting a potential immunoregulatory function for the soluble molecule.

Occupancy of the B cell glycoprotein, CD72 results in syk-independent activation of phospholipase-C gamma and calcium mobilization. The cytoplasmic tail of CD72 does not contain an immunoreceptor tyrosine-based activation motif to directly transduce signals into the B lymphocyte. Hence, we investigated whether other coreceptors such as CD19 and its associated phosphatidylinositol 3-kinase (PI 3-K) were involved in CD72 signaling. Two specific inhibitors of PI 3-K inhibited CD72-stimulated B cell proliferation in a dose-dependent manner. Activation of B lymphocytes via CD72 resulted in recruitment and activation of PI 3-K, which was mediated by CD19. Accordingly, CD72 ligation induced CD19 tyrosine phosphorylation. Thus, lipid products generated as a result of PI 3-K activation may have an important function in CD72-mediated B lymphocyte activation. The kinetics of CD19 tyrosine phosphorylation induced by CD72 ligation were strikingly different from those seen following B cell antigen receptor (BCR) stimulation. A transient increase in the tyrosine phosphorylation of the complement receptors, CD21 and CD35 was observed in BCR- but not CD72-stimulated cells. Co-cross-linking of CD72 and CD19 failed to induce syk tyrosine phosphorylation suggesting that even under these conditions, CD72 signaling was independent of syk activation. A transient and stimulation-dependent physical association between CD19 and CD72 was observed in CD72-ligated cells. These observations suggest a mechanism by which CD72 can recruit CD19 and influence activation of CD19-associated PI 3-K, which appears to be critical for CD72-mediated B cell activation.

Numerous allergens have proteolytic activities. It has been speculated that this property may contribute to their allergenicity. Therefore, we have evaluated the effect of different physiological protease inhibitors (PI) on the regulation of human IgE synthesis. Unexpectedly, the serine PI, alpha-1 antitrypsin, also called alpha-1 protease inhibitor (alpha1PI), induced a potent and selective dose-dependent increase of IgE and IgG4 production by human tonsillar B cells stimulated with the IgE and IgG4 switch factors, IL-4 and anti-CD40 mAb. The other serine PI tested were inefficient. Furthermore, this effect of alpha1PI was accompanied by an increase in (1) germ-line and mature sigma mRNA transcription, (2) proliferation and (3) membrane CD23 and CD21 expression, while the expression of other molecules involved in the regulation of IgE synthesis was unchanged. Since CD23-CD21 pairing plays a crucial role in the up-regulation of IgE synthesis, we have tested whether blocking this interaction affected alpha1PI-increased IgE production. The neutralizing anti-CD23 mAb, Mab 25, partly reversed the IgE increase caused by alpha1PI. Moreover, alpha1PI potentiation of IgE synthesis was prevented by elastase, a natural substrate of alpha1PI, thereby suggesting that alpha1PI may inhibit endogenous B cell enzyme(s) involved in the down-regulation of IgE synthesis. Alpha1PI also potentiated IgE and IgG4 production by IL-4-stimulated peripheral blood mononuclear cells but was not a switch factor for IgE and IgG4 as it was unable to replace IL-4 or anti-CD40 mAb in inducing IgE and IgG4 production. In conclusion, this study shows that alpha1PI acts as a potent co-stimulus for IgE and IgG4 synthesis and suggests that the equilibrium between protease/ protease inhibitor participates in the control of human IgE and IgG4 synthesis.

This report describes two subpopulations of B cells in sheep. These subpopulations have distinct recirculation characteristics and tissue distributions. Phenotypically the populations are distinguished by their differential expression of the complement receptors, CD21 (CR2) and CD11b/CD18 (CR3). CD11b+ B cells are surface (s)IgMhi, co-express CD11c but are L-selectin negative. They populate the splenic marginal zone but are absent from splenic and ileal Peyer's patch (IPP) follicles and both afferent and efferent lymph compartments. Fluorescent tracing experiments showed that the CD11b+ B cells are non-recirculating as they did not appear in lymph after intravenous inoculation but are restricted to the blood and spleen. The CD11b-negative population expresses a conformational determinant of CD21 that is recognized by the monoclonal antibody Du 2-74. These cells are sIgMlo and co-express L-selectin. They populate the splenic and IPP follicles, are absent from the splenic marginal zone and are the only B cells in afferent lymph, efferent lymph and all lymph nodes. Fluorescence tracing experiments showed that the CD21 B cells are recirculating cells with their entry into efferent lymph being detectable by 16 h and peaking at 24-30 h. These data suggest that there are at least two lineages of B cells in the sheep with different phenotypic, functional and recirculation characteristics.

We previously demonstrated that human C3d or pep16, a 16-amino acid synthetic peptide derived from human C3d, induced in vivo and in vitro tyrosine phosphorylation of pp105, an intracellular component found only in human cells that express CR2 at their surface. To determine the contribution of CR2 molecules to this enzymatic regulation, we first analyzed whether activation of CR2 by other extracellular CR2 ligands could trigger such regulation in cell extracts. Subsequently, we used cell extracts of either CR2-positive cells depleted in CR2 molecules by absorption with anti-CR2 antibodies or CR2-negative cells transfected with CR2 cDNA. We demonstrate here that pp105 phosphorylation was induced when CR2 was activated by C3d and pep16 as well as by gp350, the Epstein-Barr virus capsid protein or OKB7, an anti-CR2 monoclonal antibody (mAb). HB5, another anti-CR2 mAb, which did not activate B lymphocytes through CR2, did not induce pp105 phosphorylation. Thus, C3d, pep16, gp350, and OKB7 presented similar properties in activating CR2 to trigger pp105 phosphorylation and in regulating B lymphocyte proliferation, while HB-5 had no effect on either assays. Furthermore, our data demonstrate that the presence of CR2 activated by its extracellular ligands regulates pp105 phosphorylation through two distinct pathways: one which also requires the presence of non-activated CD19, and one which is independent of CD19. The involvement of CD19 in the first pathway was not due to the formation of putative CR2-CD19 complexes. Both pathways were TAPA-1 independent. This is the first demonstration that activated CR2 molecules can play a regulatory role in enzymatic function, such as phosphorylation, despite the absence of CD19 and TAPA-1.

Stable transfection of Epstein-Barr virus (EBV) nuclear antigen 2 (EBNA2) expressed as a fusion protein with the hormone-binding domain of the estrogen receptor was used to study expression of CD21 and other surface markers in different cell lines. Special emphasis was placed on cell lines with a normally low expression of CD21, especially on T cell lines. After induction of EBNA2, a substantial increase in CD21 mRNA was observed, as well as increased production of membrane CD21. This was found not only in cell lines of B cell origin, but also in the T cell line Jurkat. The amount of CD21 was quantitated by means of a fluorescence immunoassay, and found to correlate with the presence of EBNA2 protein. A decrease in EBNA2 abundance was associated with complete loss of cell-associated CD21. As we could also detect large amounts of soluble CD21 (sCD21) in the supernatant of the transfected cell lines, which exceeded the total amount contained in the respective cell lysates, this indicates considerable shedding of the newly synthesized receptor molecules induced by EBNA2, comparable to the situation described for CD23. It further provides an explanation of the recent findings of increased sCD21 levels in sera of patients with EBV-associated disease, and suggests a possible additional function of EBNA2 in vivo.

Recombinant full-length human CD23 incorporated into fluorescent liposomes was used to detect a ligand for CD23 on the basophilic leukemia cell line, KU 812. Based on our recent finding that CD23 interacts with CD21 on subsets of B and T cells, we investigated if the same ligand was involved on KU 812 cells. An anti-CD21 monoclonal antibody (mAb) BU-33, was able to totally block CD23-liposome binding to KU 812 cells. Moreover, KU 812 cells express CD21 mRNA and have a cell surface molecule that reacts with anti-CD21 mAb. The CD23/CD21 interaction was not merely physical but was also associated with an increase in histamine release by KU 812 cells. Both recombinant soluble CD23 and an anti-CD21 mAb-mediated effect on histamine release was not restricted to and anti-CD21 mAb-mediated effect on histamine release was not restricted to the leukemic cell line, but was also observed with normal human blood basophils. These data demonstrate that CD21 is expressed on basophilic cells and that CD21 controls histamine production upon ligand-induced stimulation (CD23 or anti-CD21 mAb).

We have previously found that interleukin-4 and CD40 monoclonal antibodies (mAb) are strong potentiators of homotypic B cell aggregation which is dependent on LFA-1. We show here that CD23 mAb were also able to inhibit aggregation to a similar extent as LFA-1 antibodies. This inhibition was restricted to the MHM6 epitope of CD23 and antibodies to other epitopes [Epstein-Barr virus (EBV) CS-1, EBV CS-2, EBV CS-5 and mAb 25] or occupation of the Fc-binding site by IgE had no or a slightly enhancing effect on aggregation. When testing two antibodies to CD21, the recently defined ligand for CD23, one of these (BU32) was found to be inhibitory whereas the other (THB5) had no effect. By combining antibodies to LFA-1 and CD23, aggregation was often completely inhibited. These data suggest that LFA-1/ICAM-1 and CD23/CD21 are the major molecules involved in homotypic aggregation of human B cells.

On a panel of nine human B cell lines we showed that the expression of the complement regulatory factors complement receptor type 2 (CR2; CD21), decay-accelerating factor, (DAF; CD55) and homologous restriction factor (HRF20, CD59) is not correlated. All lines expressed DAF, six lines carried detectable amounts of CR2 and three carried HRF20. Upon incubation in human serum, under conditions which allowed the activation of complement through the alternative pathway, the CR2-carrying lines bound C3 fragments and two of them (Ramos and one of its two sublines) were damaged. These two lines had the lowest DAF expression, less than 50% of the cells reacted with the IA10 monoclonal antibody. By modulating the expression of the complement regulatory molecules, the lytic sensitivity of the B cell lines could be altered. Blockade of DAF on the HRF20-, CR2+ lines with the specific monoclonal antibodies increased their sensitivity to lysis by human serum. With the DAF- and HRF20+ cells significant lytic effect was obtained only when they were pretreated with both of the specific antibodies. Interferon-gamma or tumor necrosis factor-alpha treatment elevated the amount of CR2 on the low-CR2 expressor line (Ramos/HR1K) which thereafter bound higher amounts of C3 fragments and was lysed when incubated in human serum. This line had relatively low DAF level and lacked HRF20. The cytokine treatment did not alter the expression of these molecules. The CR2+ Ramos and the CR2- Rael cells were treated with 5-azacytidine which induced HRF20 and increased DAF expression. In parallel with this change Ramos cells became resistant to C-mediated lysis. The experiments with the panel of human B cell lines showed thus that cytolysis through activation of complement in homologous serum can be regulated at several steps by cell surface molecules. While expression of CR2 was required for C3 fixation, DAF and HRF20 inhibited lysis. By independent modulation of the quantities of these molecules, cells acquired or lost their sensitivity.

Multivalent but not monovalent CR2 ligands are required to elicit Raji cell proliferation as well as other B cell responses. It has been reported (C. Servis and J. D. Lambris, J. Immunol. 1989. 142: 2207) that the tetrameric peptide T-(C31202-1214)4, which represents the CR2-binding site in C3d, was able to support Raji cell growth. We show here that the tetrameric peptide T-(gp350(19-30)4, which contains the CR2-binding site in gp350 protein of EBV also induces Raji cell growth and this effect is inhibited by the monomeric peptides gp350(19-30) and C3(1201-1214). We also investigated the nature of the interaction between C3 fragment and CR2 in order to explain the Raji cell growth-supporting effect exerted by C3. The following findings suggest that there are multiple sites in the C3 molecule able to interact with CR2: (1) both C3c and C3d immobilized on microspheres are able to bind to Raji cells through CR2. (2) soluble C3d inhibits to a greater extent the binding of CR2 to fixed C3d than to fixed C3b, which suggests the existence of additional CR2-binding sites within C3b not present in the C3d portion of the molecule; (3) synthetic peptides C3(1187-1214), C3(741-757) and C3(295-307) which represents regions of similarity in the C3 molecule bind specifically to CR2 on Raji cells and compete with each other for binding to the receptor and (4) preincubation of microtiter plate-fixed C3b with monoclonal or polyclonal anti-peptide antibodies (C3-9, anti-C3(727-768) recognize the N terminus of the alpha chain of C3 (including residues 741-757) inhibited CR2 binding. Therefore, these data suggest that the N terminus of the alpha chain of C3 is involved in binding to CR2.

The importance of the complement system for mounting an antibody response in vivo was investigated by down-regulating and blocking the complement receptors (CR) in mice with three different monoclonal rat antibodies (mAb): mAb 8C12 recognized the C3b-binding site of CR1, mAb 7G6 recognized another site of CR1 and the C3d-binding site of CR2 and mAb 7E9 again recognized other epitopes on CR1 and CR2. We have earlier shown that 7G6, is the only mAb that completely suppresses the primary antibody response to horse erythrocytes. This antibody was also shown to suppress induction of immunological memory and a secondary, antibody response. In contrast to what seems to be the case for thymus-dependent antibody responses, all three mAb could inhibit the antibody response to a thymus-independent antigen, dextran B 1355S.

Membrane regulatory molecules normally prevent complement activation by autologous cells, therefore we compared the membrane control system of human lymphoid cell lines which activate or not human complement through the alternative pathway (AP). Membrane expression of decay-accelerating factor (DAF), membrane cofactor protein (MCP), complement receptors (CR)1, CR2 and H was measured either by radioimmunoassay or enzyme-linked immuno-sorbent assay on cell lysates. Soluble extracts of isolated membranes were tested functionally for their ability to accelerate the decay of C3bBb C3-convertase and allow the cleavage of C3b by factor I. Both regulatory functions were detected in solubilized membranes of Ramos cells, which do not activate the AP, as well as on the potent AP activator, Raji. Raji cells were found to express CR2, DAF and MCP molecules, while MCP was the only known regulatory protein detected on Ramos cells which expressed neither CR1, nor CR2, H or DAF. The I-cofactor activity of both Raji and Ramos cells was immunoprecipitated by anti-MCP, but the decay-accelerating activity was not adsorbed by anti-DAF nor by any of the available antibodies. Two EBV genome-negative cell lines (BJAB, BL41) were tested before and after in vitro conversion by EBV. As previously shown, EBV-converted cell lines activate the AP more efficiently than EBV− cell lines. At the same time, EBV superinfection induces an increase of both AP regulatory functions of cell membranes and enhances the expression of DAF, MCP and CR2. The results of this study show that complement activation by lymphoid cell lines is not related to an impaired autologous control of these cells, but that the expression of regulatory molecules increases together with the appearance of activating structures on the cell surface. Our results also suggest the occurrence of a new factor involved in the decay-accelerating activity on BL lines.

The Epstein-Barr virus (EBV) and the C3d component of complement bind to the same cell surface receptor (EBVR/CR2) which is part of the B lymphocyte differentiation antigen recognized by the monoclonal antibodies (mAb) of the cluster of differentiation 21 (CD21). To analyze EBV and C3d binding to this receptor, mouse fibroblasts were transfected with human genomic DNA and rare CD21-positive cells were selected and cloned by cell sorting. The presence of the human gene in host cell DNA as well as its transcription product were assayed with a cloned EBVR/CR2 cDNA by Southern and Northern blotting analysis, respectively. A glycoprotein of apparent molecular mass of 140 kDa, similar to that found in human B lymphocytes, was immunoprecipitated with anti-CD21 mAb and proved to be functional since both C3d and EBV bound efficiently and specifically to mouse cells expressing EBVR/CR2. However, no expression of EBV nuclear antigens, early antigens and viral capsid antigens was detected in cells exposed to EBV. This indicates that the presence of EBVR/CR2 is not sufficient to allow full infection of mouse fibroblasts.

gp140, the C3d/EBV receptor (CR2), previously isolated and characterized from human B lymphocytes, was identified on human platelets: (a) by measuring the specific binding of either polyclonal anti-gp140 IgG and monoclonal anti-C3d/EBVR antibodies, as OKB-7 and HB-5, or human C3d; (b) by isolating gp140 from solubilized platelet components with polyclonal anti-gp140 IgG or monoclonal OKB-7, using immunoprecipitation and electro-immunoblotting assays; (c) by inducing specific activation of human platelets. Cross-linking of this receptor by polyclonal anti-gp140 IgG induced aggregation of human platelets and stimulated ATP release. Absence of lactate dehydrogenase release and inhibition by EDTA and prostacyclin of anti-gp140-induced aggregation, support strongly active aggregation and absence of lysis. Platelet aggregation by anti-gp140 required metabolic activities and was modulated by fibrinogen, paf-acether or thrombin. OKB-7 triggered human platelet aggregation when cross-linked by anti-mouse second-step antibodies. In the same way, platelet activation by C3d fragment was detected, in presence of fibrinogen, only when C3d was cross-linked on the cell surface by anti-C3d F(ab′)2 fragments.

The C3d receptor is a specific marker of B lymphocytes. Recently we have shown that C3d receptor activity is carried by a gp 140 membrane antigen. A polyclonal antibody has been prepared by immunizing a rabbit with highly purified gp 140 molecule isolated from membranes of the human B lymphoblastoid cell line Raji and its high specificity was previously demonstrated. We tested the effect of this antibody to the C3d receptor on the B cell proliferative response. Purified B cells from human blood were induced to proliferate by a B cell growth factor (BCGF)-containing partially purified supernatant from activated T cells. The anti-C3d receptor F(ab')2 enhanced the BCGF-dependent B cell proliferation. This effect was dose dependent, was observed in the presence of different concentrations of BCGF and did not correspond to a change in the time course of the response. The anti-C3d receptor F(ab')2 had no mitogenic effect in the absence of T cell supernatant. In contrast the undigested anti-C3d receptor IgG suppressed the BCGF-dependent B cell proliferation. These results emphasize the potentialities of anti-gp 140 F(ab')2 to explore the involvement of the C3d receptor in the regulation of B cell response to T cell products.

In the human spleen the marginal zone (MZ) contains medium-sized B cells with a distinct immunophenotype. A main function attributed to the MZ is its involvement in the primary response to blood-borne antigens, in particular to thymus-independent antigens type 2. In this study the presence of antigens related to activation and proliferation was evaluated in human spleens by immunohistochemical staining. It appeared that MZ B cells do not show interleukin 2 receptor expression, and are in G0 phase of the cell cycle, as demonstrated by the lack of Ki-67 reactivity. The most interesting finding is the high CD21 expression by MZ B cells in the absence of IgD expression. As the CD21 antigen has been shown to be involved in B cell activation in close linkage with IgM, it can be suggested that MZ B cells are particularly well equipped for rapid and easy activation in a primary immune response.

gp140, the EB/C3d receptor (EBV/C3dR; CR2), is a membrane site involved in human B cell regulation. Cross-linking of this receptor on the cell surface by its specific ligands led to the enhancement of B cell proliferation in synergy with T cell factors. In vitro activation of human peripheral B lymphocytes by cross-linking membrane immunoglobulins with anti-mu antibody induced EBV/C3dR phosphorylation. These studies were pursued by analyzing cell-free phosphorylation of EBV/C3dR isolated from Raji cell fractions, and immobilized on OKB7, a monoclonal anti-EBV/C3dR antibody. Three EBV/C3dR-related antigens which could be cell-free phosphorylated were detected: gp140, the EBV/C3dR, p130 and p120. gp140, the mature form of EBV/C3dR, was isolated from plasma membrane and from purified nuclei. p130 was identified as an intracellular intermediate of EBV/C3dR glycosylation, localized in low-density microsomes. Phosphoamino acid analysis of EBV/C3dR allowed the detection of phosphotyrosine and phosphoserine residues. These data suggest that EBV/C3dR could carry an autophosphorylation activity and could be associated to serine kinases. Using polyclonal anti-p120 antibody and anti-120 kDa nuclear ribonucleoprotein monoclonal antibody (mAb), p120 was identified as a nuclear ribonucleoprotein antigenically not related to EBV/C3dR. Detection of p120 on EBV/C3dR, immobilized on OKB7, was due to interactions between both antigens, instead of anti-EBV/C3dR mAb cross-reactivity with p120. Cell-free phosphorylation of p120 was under the control of EBV/C3dR. However, it is not yet established whether other nuclear or membrane components were involved in the control of p120 cell-free phosphorylation by EBV/C3dR. From the data presented herein, we propose that phosphorylation of a 120-kDa nuclear ribonucleoprotein by EBV/C3dR-associated kinases could represent a crucial step in in vivo regulation of human B cell activation.

Gp140 is the Epstein-Barr virus receptor and the C3d receptor (EBVR/C3dR) of human B lymphocytes. Recently, we have shown that cross-linking of EBVR/C3dR on cell surface by polyclonal anti-gp140 induced B cell activation, in presence of T cell factors. Immunoregulatory abnormalities of EBV-induced B cell activation have been demonstrated in rheumatoid arthritis (RA) patients. These data prompted us to analyze the putative presence of anti-EBVR/C3dR autoantibodies in human sera. The IgG fractions from eleven RA and 10 normal sera were tested for their ability to: (a) bind to Raji cell surface; (b) inhibit the binding to cell surface of 3 anti-EBVR/C3dR monoclonal antibodies (mAb), which recognized different epitopes on gp140; (c) inhibit the binding of particle-bound C3d and (d) react with 1% Nonidet-P40-solubilized gp140 from Raji cell membranes, in immunoblotting assays. Three RA sera carry anti- EBVR/C3dR autoantibodies which react with gp140 expressed on Raji cell surface or its solubilized form. The purification of monomeric IgG fraction of selected RA sera ruled out involvement of immune complexes carrying C3 molecules, which could interfere in these assays. One of these 3 RA sera was able to inhibit the binding to cell surface of anti-EBVR/C3dR mAb and particle-bound C3d. However, the 2 other RA sera, found positive by immunoblotting, did not inhibit particle-bound C3d and presented differences in their inhibitory effect on anti-EBVR/C3dR mAb binding to Raji cell surface. These data allow us to demonstrate differences which exist in the properties of anti-EBVR/C3dR autoantibodies. These autoantibodies were not detected in all the normal and other RA sera. Anti-EBVR/C3dR autoantibodies could play a role “in vivo” in B lymphocyte activation of RA patients.

The biologic effects of the third component of complement, C3, are mediated via receptors which specifically bind the enzymatic degradation products resulting from the cleavage of C3. One of the products, C3d, has been associated with binding to the second complement receptor CR2 (CD21). This receptor, which is identical to the receptor for Epstein-Barr virus (EBV), has been primarily found on cells of the B lineage, but not on mature T cells or other cells of erythroid or myeloid lineages. In the present investigation, we report the presence of CR2 on human thymocytes. Indirect immunofluorescence analysis employing monoclonal anti-CR2 antibodies revealed a range of thymocyte reactivity from 15% to 63% in thirteen experiments using cells of different donors. Reactivity was always greater with the monoclonal anti-CR2 (CD21) antibody HB-5 than with two other antibodies which recognize distinct epitopes on the CR2 molecule. Two-color immunofluorescence analysis indicated that the brightest of the HB-5-stained thymocytes also reacted with the monoclonal anti-CD1 antibody T6 (immature thymocyte marker) while some of the duller HB-5-staining cells reacted with the monoclonal anti-CD3 antibody Leu-4 (mature thymocyte marker). Immunoprecipitation of CR2 on thymocytes with antibody HB-5 and polyacrylamide gel electrophoretic analysis revealed a protein of 145 kDa molecular mass which is consistent with the size of CR2 found on B lymphocytes. These findings raise several questions regarding the biologic role of CR2-EBV receptor on cells of the T lineage.

Epstein-Barr virus (EBV) infects and transforms human B lymphocytes. The virus receptor was shown to be identical to the complement receptor CR2. The consequences of EBV/CR2 receptor interaction on B lymphocyte activation were analyzed by infection of B cells with the transforming (B95-8) and the nontransforming (P3HR1) virus strains and with UV-inactivated B95-8 virus. Similar to mitogens and antibodies against surface IgM and CR2 receptor, transforming and nontransforming EBV induced the release of leukocyte migration inhibitory factor (LIF). LIF production occurred early after infection and was not affected by irradiation of the B95-8 virus with doses of UV-light which prevented the expression of viral functions. B lymphocytes infected with UV-inactivated virus responded to T cell-derived B cell growth factors. The results demonstrate that binding of EBV to the CR2 receptor activates the B cells. Progression of the infected cells in the activation pathway required helper cell-derived factors or signals provided by the intact viral genome. The nontransforming P3HR1 virus induced a low level of RNA synthesis and increase of cell size and major histocompatibility complex class I antigen expression. Only the transforming virus induced expression of EBV nuclear antigens and cellular DNA synthesis.

Normal human B cell proliferation is controlled by various immunoregulatory signals including the T cell-derived lymphokine B cell growth factor (BCGF). Human BCGF provides the final proliferative signal to normal, activated B cells. We herein show that anti-CR2 monoclonal antibodies inhibit human B cell responsiveness to purified BCGF. Addition of anti-CR2 antibody, AB5, was capable of completely inhibiting BCGF-mediated enhancement of either anti-mu or staphylococcal protein A-activated human B cells (191 +/- 21 cpm vs. 3942 +/- 622 cpm, mean +/- SEM). Inhibition of B cell response to BCGF by AB5 occurred in a dose-dependent manner. Monoclonal antibody anti-B2, which recognizes the same 140-kDa glycoprotein as AB5, in comparable concentrations also inhibited B cell responsiveness to BCGF. Monoclonal antibodies of the same subclass (IgG1) showed no inhibitory effect on BCGF enhancement of B cell proliferation. The F(ab')2 fragment of AB5 generated by pepsin digestion was similarly inhibitory as was the intact Ig. AB5-mediated inhibition was independent of the target B cell state of activation. Both resting and activated B cells (anti-mu or staphylococcal protein A activated) incubated with similar concentrations of AB5 were unresponsive to BCGF. The ability of anti-CR2 antibodies to block BCGF-dependent B cell proliferation suggests that occupancy of C3d membrane receptors may result in modulation of B cell proliferation in physiologic or clinical disease states.